如果我们使用matplotlib.pyplotsubplot，那么我们可以迭代get_xticks()，并使用list-comprehension索引agesx中的标签，以获得_xticklabels和set，list作为xticklabels。我们不想改变xticks，因为第0个xtick应该是18，我们想改变的是xticklabels，并保持xticks不变：

import matplotlib as mpl
from matplotlib import pyplot as plt
import numpy as np

agesx = [18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35,
          36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55]

py_devy = [20046, 17100, 20000, 24744, 30500, 37732, 41247, 45372, 48876, 53850, 57287, 63016, 65998, 70003, 70000, 71496, 75370, 83640, 84666,
            84392, 78254, 85000, 87038, 91991, 100000, 94796, 97962, 93302, 99240, 102736, 112285, 100771, 104708, 108423, 101407, 112542, 122870, 120000]

js_devy = [16446, 16791, 18942, 21780, 25704, 29000, 34372, 37810, 43515, 46823, 49293, 53437, 56373, 62375, 66674, 68745, 68746, 74583, 79000,
            78508, 79996, 80403, 83820, 88833, 91660, 87892, 96243, 90000, 99313, 91660, 102264, 100000, 100000, 91660, 99240, 108000, 105000, 104000]

all_devy = [17784, 16500, 18012, 20628, 25206, 30252, 34368, 38496, 42000, 46752, 49320, 53200, 56000, 62316, 64928, 67317, 68748, 73752, 77232,
         78000, 78508, 79536, 82488, 88935, 90000, 90056, 95000, 90000, 91633, 91660, 98150, 98964, 100000, 98988, 100000, 108923, 105000, 103117]

width = 0.25
x_indexes = np.arange(len(agesx))
#plt.xticks(ticks=x_indexes,labels=agesx)

ax = plt.subplot(1, 1, 1)
plt.style.use("seaborn-dark")
ax.bar(x_indexes-width, all_devy, width=width, label= "All Devs")
ax.bar(x_indexes,py_devy,width=width, label= "Python Devs")
ax.bar(x_indexes + width, js_devy, width=width, label= "Javascript Dev")

ax.set_title("Median Developer Salaries by Age (USD)")
ax.set_xlabel("Ages")
ax.set_ylabel("Salary (USD)")
plt.tight_layout()
_xticklabels = [agesx[int(_xtick)] for _xtick in ax.get_xticks() if int(_xtick) < len(agesx)]
ax.set_xticklabels(_xticklabels)
ax.legend()
plt.show()

输出：

如果我们希望xticklabels在5和10中，同时从18开始，那么我们还必须修改xticks并使用numpy的arange函数，使其成为list，insert成为0的索引：

...
_xticks = list(np.arange(2, len(agesx), 5).astype(int))
_xticks.insert(0, 0)
ax.set_xticks(_xticks)
_xticklabels = [agesx[_xtick] for _xtick in ax.get_xticks()]
ax.set_xticklabels(_xticklabels)
...

输出：

或者如果我们想要每隔一个agesx标签：

...
_xticks = np.arange(0, len(agesx), 2).astype(int)
ax.set_xticks(_xticks)
xticklabels = [agesx[_xtick] for _xtick in ax.get_xticks()]
ax.set_xticklabels(xticklabels)
...

输出：

但是我们不能设置agesx的每个标签，否则会有重叠的xticklabels：

...
_xticks = np.arange(0, len(agesx), 1).astype(int)
ax.set_xticks(_xticks)
xticklabels = [agesx[_xtick] for _xtick in ax.get_xticks()]
ax.set_xticklabels(xticklabels)
...

输出：

• 问题使用隐式的“pyplot”接口，因此，两个简单的选项是：
• 增加图形的宽度：
• fig = plt.figure(figsize=(10, 6))
• plt.figure(figsize=(10, 6))。
• 使用plt.xticks设置刻度数
  *在python 3.11.2、pandas 2.0.1、matplotlib 3.7.1中测试

增大图幅

# increase the figure size
plt.figure(figsize=(10, 6))

plt.bar(x_indexes-width, all_devy, width=width, label= "All Devs")
plt.bar(x_indexes,py_devy,width=width, label= "Python Devs")
plt.bar(x_indexes + width, js_devy, width=width, label= "Javascript Dev")
_ = plt.xticks(ticks=x_indexes, labels=agesx)

指定xtick标签plt.xticks

• plt.xticks(np.arange(18, 55, 3))不起作用，因为与线图不同，条形图的刻度是0索引的。print(plt.gca().get_xticklabels())在索引0处显示'18'。
• 使用plt.xticks(ticks=x_indexes, labels=agesx)预设刻度和标签
• [v if i%2 == 0 else '' for i, v in enumerate(agesx)]使用列表解析将每隔一个标签设置为空字符串。
• 增加2，以减少刻度标签。

plt.bar(x_indexes-width, all_devy, width=width, label= "All Devs")
plt.bar(x_indexes,py_devy,width=width, label= "Python Devs")
plt.bar(x_indexes + width, js_devy, width=width, label= "Javascript Dev")

# adjust the frequency of the visible labels; the number of ticks and labels must be the same
_ = plt.xticks(ticks=range(len(agesx)), labels=[v if i%2 == 0 else '' for i, v in enumerate(agesx)])

• 或者，使用pandas，这使得分组和堆叠条形图更容易。matplotlib是pandas.DataFrame.plot的默认后端。
• 下面的示例将返回的matplotlib.axes.Axes分配给ax，这是显式的“Axes”接口。

import pandas as pd

# load data to pandas
data = {'Python Devs': py_devy, 'Javascript Devs': js_devy, 'All Devs': all_devy}
df = pd.DataFrame(data=data, index=agesx)

# 1. plot the bars with rotated labels
ax = df.plot(kind='bar', width=0.85, rot=90)

# 2. plot with increased figure size
ax = df.plot(kind='bar', width=0.85, figsize=(10, 6), rot=0)

# 3. plot horizontal bars, which may be better for many bars, and long labels
ax = df.plot(kind='barh', width=0.85, figsize=(6, 7))

# 4. every other tick label
ax = df.plot(kind='bar', width=0.85, rot=0, xticks=range(len(df.index))[::2])
_ = ax.set_xticklabels(df.index[::2])

plot1plot2plot3plot4
df.head()

Python Devs  Javascript Devs  All Devs
18        20046            16446     17784
19        17100            16791     16500
20        20000            18942     18012
21        24744            21780     20628
22        30500            25704     25206

下面的条形图在x轴上太杂乱了。
IMHO混乱是Matplotlib试图告诉你，你的图是相当拥挤的。

import matplotlib.pyplot as plt
import numpy as np

agesx = [18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55]
py_devy =  [20046, 17100, 20000, 24744, 30500, 37732, 41247, 45372, 48876, 53850, 57287, 63016, 65998, 70003, 70000, 71496, 75370, 83640, 84666, 84392, 78254, 85000, 87038, 91991, 100000, 94796, 97962, 93302, 99240, 102736, 112285, 100771, 104708, 108423, 101407, 112542, 122870, 120000]
js_devy =  [16446, 16791, 18942, 21780, 25704, 29000, 34372, 37810, 43515, 46823, 49293, 53437, 56373, 62375, 66674, 68745, 68746, 74583, 79000, 78508, 79996, 80403, 83820, 88833, 91660, 87892, 96243, 90000, 99313, 91660, 102264, 100000, 100000, 91660, 99240, 108000, 105000, 104000]
all_devy = [17784, 16500, 18012, 20628, 25206, 30252, 34368, 38496, 42000, 46752, 49320, 53200, 56000, 62316, 64928, 67317, 68748, 73752, 77232, 78000, 78508, 79536, 82488, 88935, 90000, 90056, 95000, 90000, 91633, 91660, 98150, 98964, 100000, 98988, 100000, 108923, 105000, 103117]

rmean = lambda l,n:[sum(t)/n for t in zip(*(l[i:]for i in range(n)))]
N = 5

plt.style.use("seaborn-v0_8-dark")
for y, label in zip((py_devy,js_devy,all_devy),
                  'Python JavaScript all'.split()):
     plt.plot(rmean(agesx, N), rmean(y, N), label=label+' devs')
plt.ylim((-1000, 120000))
plt.grid()
plt.legend()
plt.xlabel('Central Age for the running mean')
plt.ylabel('Median Salary, €')
plt.suptitle('              Median Developer Salary')
plt.title('— running mean over %d age classes —'%N, size='small')
plt.tight_layout()
plt.show()